Announcement

Collapse
No announcement yet.

Other Open-Source / Linux Letdowns For 2018 From File Creation Time To Flatpaks

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • #21
    The title of this article seems like total click-bait.
    Flatpak works great on an updated Fedora 28+. Flathub is great. Things definitely moved forward strongly during the year to the point that almost nobody knew what Flatpak was (as it wasn't very useful) at the beginning of the year, and by the time of Fedora 29's release, it was a full solution ready for prime time. By now it's clearly the way of the future, especially as we move to "atomic" root filesystems, like Android, but even without that, it's still adding nicely to overall user experience, making the better, safer-way easier to use than the older, less secure way. Win-win!

    Comment


    • #22
      Originally posted by jo-erlend View Post
      For Ubuntu, it was easy; we needed a stable, mature and modern desktop by April 2014. Look at the current status of Gnome Shell and tell me Gnome Shell was the best bet in 2010 in order to achieve that goal.
      look at the status of compis in 2010 and compare it to current gnome shell. surely you could've improved gnome shell from 2010 to 2014 just as you've improved compis
      Originally posted by jo-erlend View Post
      These days, however, close to a decade later, Gnome is about to become a good foundation for a modern desktop. But sometimes, waiting a decade and hoping that someone else will fix the problem, is not the best solution. Sometimes you just need to make the stuff you require.
      yes, someone has to do it alone like you. and to be laughed at afterwards

      Comment


      • #23
        Originally posted by computerquip View Post
        As for Mir and Snapcraft, that was pure NIH.
        not really. as usual that was canonical trying to do vendor lock-in. which is evil, unlike nih

        Comment


        • #24
          Originally posted by Weasel View Post
          It doesn't need users. It needs a sane development environment and userland.
          who will develop commercial software for no users?

          Comment


          • #25
            Originally posted by jo-erlend View Post

            Isn't this a very strange argument? Unity is based on the long-existing Compiz, whereas Gnome decided to create a brand new compositor. Wouldn't it be Gnome that should've just built their desktop on existing technologies, as you say?

            For Ubuntu, it was easy; we needed a stable, mature and modern desktop by April 2014. Look at the current status of Gnome Shell and tell me Gnome Shell was the best bet in 2010 in order to achieve that goal.

            These days, however, close to a decade later, Gnome is about to become a good foundation for a modern desktop. But sometimes, waiting a decade and hoping that someone else will fix the problem, is not the best solution. Sometimes you just need to make the stuff you require.
            From that point of view, it's true. However, I meant to have lesser diversity of base DE, and customize it on the top most level.

            You're right. I loved Gnome 2 and KDE 3.5+. When distributions started shipping only non usable early versions Gnome 3 and KDE 4, then if was a bit of pain.. Until Gnome 3.18 which was released, which was finally usable.

            Comment


            • #26
              Originally posted by Weasel View Post
              It doesn't need users. It needs a sane development environment and userland.
              Considering we're talking about software developed by private companies who make their money selling their software to end users, not something worked on by hobbyists in their spare time, users are absolutely necessary. Saying what you're saying is like saying that you don't need water to go boating, you need solid timber and good tools.

              Comment


              • #27
                To the Canonical guy saying gnome-shell wasn't ready, honestly, I've been using it daily since that first experimental version on Launchpad, in the time when Zeitgeist was still a thing in Ubuntu and Unity wasn't even a consideration, and considered it good/stable/better than whatever else was available already at that time. The announcement that Ubuntu was abandoning it to rollout their own solution is exactly what made me leave Ubuntu for Fedora for good. I never looked back.

                Comment


                • #28
                  Originally posted by pal666 View Post
                  who will develop commercial software for no users?
                  Originally posted by L_A_G View Post

                  Considering we're talking about software developed by private companies who make their money selling their software to end users, not something worked on by hobbyists in their spare time, users are absolutely necessary. Saying what you're saying is like saying that you don't need water to go boating, you need solid timber and good tools.
                  Sigh, you guys are missing the point. MS-DOS and Windows did not have users at the beginning either, neither did Linux for that matter.

                  The point is that: if you build it, they will come. Obviously it's not just users who matter but also the resources needed to port and maintain it. A sane development environment vastly eases out the latter. Just look how many huge companies make such crap software in Windows land just to cut costs and rarely update something unless it breaks (which is quite rare in Windows unless it's broken). Getting a new fucking ABI-breaking library every month like on Linux massively increases maintenance burden and costs.

                  Most users won't come to an OS without a sane development environment, even if they don't care about it directly (because they're not devs), they will complain about "X update broke Y app" which entails this without them realizing obviously.

                  They come to expect to download software from the vendor and install it in 2 clicks (or just run it portable) and have it work on any OS/distro instead of having to fiddle around with crap. Binary not source code. Period.

                  "But flatpak/snap aims to do that". Guess what? flatpak is perfect proof that a sane development and distribution environment is needed -- its mere existence is testament to that fact. Nobody would put so much effort into it if it wasn't needed.

                  And now I had to explain a, what I thought was obvious, one-liner.

                  Comment


                  • #29
                    My biggest letdown was hearing that x32 was losing support. On paper it seems like it would be a great architecture for what most of us want to use desktops for. I've been going through a distro hop phase lately, finally settled on Gentoo, and my first idea was to use an x32 base for the system and do x86_64 or x86 in VMs or sandboxed chroots...the next morning I came to Phoronix and decided that wasn't a good idea...just gonna go with a pure64 17.1 system once I finally finalize my zpool layout.

                    You have to do ZFS right the first time if you want optimal performance, like everything before you install the OS because once the data is written, it has to me moved off pool and back for any new settings to take so I'm doing a shitton of research. It needs special VM pools, special & multiple database pools because different DB formats like different ZFS settings, /boot done just right, /var with its own settings, /usr is a clusterfuck if you want to isolate system, user, and portage from one another for snapshot reasons...and portage needs its own special settings too. I'll end up with a LUKS encrypted ZFS /boot on USB and a native ZFS encrypted everything else for / on a two disk stripe....grub doesn't support ZFS encryption (or LUKS2) so that's one of the few ways to 100% encrypt an old bios system...it ain't fun to do right, but it'll be worth it.

                    Going that hardcore with BTRFS is just as fun...oops, didn't prevent yada from under /boot from being compressed and now it won't even boot (yes, that's a thing)...

                    Comment


                    • #30
                      Originally posted by Weasel View Post
                      Sigh, you guys are missing the point. MS-DOS and Windows did not have users at the beginning either, neither did Linux for that matter.
                      MS DOS became successful by riding on the reputation of IBM back in the day when people were still actually saying that "Nobody ever got fired for buying IBM computers". Pure and simple. Steve Ballmer and the rest of Microsoft's management at the time have been completely open about it for decades. They even had a term for it internally, referring to the experience as "Riding the Bear" as IBM was infamous for being a very demanding customer, demanding things like very heavy secrecy, refusing to pay per-copy royalties and being very cheap with the bulk sum royalties.

                      Originally the PC was supposed to run an OS called CP/M, which was superior to DOS at the time and the DOS/Windows of it's time, but it's developers had the leverage to get per-copy royalties rather than a paltry lump sum like Microsoft agreed to with DOS. Thus Microsoft made it a paid extra before ultimately dropping it altogether because the OS wasn't what people were buying the computers, it was the blue letters on the box and the user applications it ran. Just the fact that the PC was an IBM product attracted huge amounts of customers and software developers from the day it was announced and it didn't really matter that it went up against several superior and much cheaper alternatives.

                      Reputations like IBM's in the very early 1980s simply don't exist today so to get mass market appeal you absolutely need the software people want to use. Like with the competitors to the DOS PC, it doesn't really matter if you have a technically superior product.

                      Comment

                      Working...
                      X